help(gold)Description:
Daily morning gold prices in US dollars. 1 January 1985 – 31 March 1989.
Usage:
gold
Format:
Time series data
Examples:
frequency(gold)Frequency: 1
which.max(gold)770
tsdisplay(gold)autoplot(gold)help(woolyrnq)Description:
Quarterly production of woollen yarn in Australia: tonnes. Mar 1965 – Sep 1994.
Usage:
woolyrnq
Format:
Time series data
Examples:
frequency(woolyrnq)Frequency: 4
tsdisplay(woolyrnq)autoplot(woolyrnq)help(gas)Description:
Daily morning gas prices in US dollars. 1 January 1985 – 31 March 1989.
Usage:
gas
Format:
Time series data
Examples:
frequency(gas)Frequency: 12
tsdisplay(gas)autoplot(gas)First Five Values: 50.4, 49.9, 48, 48.6, 51.3
autoplot()ggseasonplot()ggsubseriesplot()gglagplot()ggAcf()autoplot( )ggseasonplot()ggsubseriesplot()ggAcf()gglagplot()Can you spot any seasonality, cyclicity and trend?
Seasonality: There is obvious seasonality apparent even in the autoplot, but this is confirmed in the ggsubseriesplot() and ggseasonplot(). There are distinct areas of high and low values that are definitely periodic in nature around the 12 month period, a yearly cycle.
Cyclicity: Cyclicity is harder to nail down, although you can clearly see in the autoplot() and ggsubseriesplot() that some of the sub-series are rising and falling for a number of seasons without relation to the season itself.
Trend:
There is no clear singular trend which is obvious from any of these plots. There are cycles which reflect micro-trends,which you can see in both theautoplot() and the ggsubseriesplot() but not a singular continuous trend throughout the entire data. It may be that there was an upward trend happening which was interupted by an external event forcing the cycles we are seeing, which otherwise might have been an upward trend.
note The lag plots are very hard to read due to so many periods and years of data. It does appear that at the 12 position the data are the most concentrated and linear, but it is not easy to read in this particular case
What do you learn about the series?
This is clearly a series of data points with notable seasonality and possible tendencies toward a trend, however the trend my be interupted by periods of market instability or extreme changes in consumer behavior toward printed materials.